相关论文: Stealing AI Model Weights Through Covert Communica…

The paper presents a novel approach of spoofing wireless signals by using a general adversarial network (GAN) to generate and transmit synthetic signals that cannot be reliably distinguished from intended signals. It is of paramount…

Collaborative machine learning settings like federated learning can be susceptible to adversarial interference and attacks. One class of such attacks is termed model inversion attacks, characterised by the adversary reverse-engineering the…

In this paper, we show that attackers can exfiltrate data from air-gapped computers via Wi-Fi signals. Malware in a compromised air-gapped computer can generate signals in the Wi-Fi frequency bands. The signals are generated through the…

Deep learning has transformed AI applications but faces critical security challenges, including adversarial attacks, data poisoning, model theft, and privacy leakage. This survey examines these vulnerabilities, detailing their mechanisms…

Machine learning models are vulnerable to simple model stealing attacks if the adversary can obtain output labels for chosen inputs. To protect against these attacks, it has been proposed to limit the information provided to the adversary…

With the rapid expansion of data lakes storing health data and hosting AI algorithms, a prominent concern arises: how safe is it to export machine learning models from these data lakes? In particular, deep network models, widely used for…

Neural networks can conceal malicious Trojan backdoors that allow a trigger to covertly change the model behavior. Detecting signs of these backdoors, particularly without access to any triggered data, is the subject of ongoing research and…

Malicious attackers can generate targeted adversarial examples by imposing tiny noises, forcing neural networks to produce specific incorrect outputs. With cross-model transferability, network models remain vulnerable even in black-box…

Neural networks are vulnerable to adversarial attacks -- small visually imperceptible crafted noise which when added to the input drastically changes the output. The most effective method of defending against these adversarial attacks is to…

AI systems can take harmful actions and are highly vulnerable to adversarial attacks. We present an approach, inspired by recent advances in representation engineering, that interrupts the models as they respond with harmful outputs with…

The use of machine learning (ML) has become increasingly prevalent in various domains, highlighting the importance of understanding and ensuring its safety. One pressing concern is the vulnerability of ML applications to model stealing…

Model stealing attack is increasingly threatening the confidentiality of machine learning models deployed in the cloud. Recent studies reveal that adversaries can exploit data synthesis techniques to steal machine learning models even in…

AI control protocols serve as a defense mechanism to stop untrusted LLM agents from causing harm in autonomous settings. Prior work treats this as a security problem, stress testing with exploits that use the deployment context to subtly…

Model merging has emerged as a promising approach for updating large language models (LLMs) by integrating multiple domain-specific models into a cross-domain merged model. Despite its utility and plug-and-play nature, unmonitored mergers…

Transfer-based attacks pose a significant threat to real-world applications by directly targeting victim models with adversarial examples generated on surrogate models. While numerous approaches have been proposed to enhance adversarial…

Training state-of-the-art (SOTA) deep learning models requires a large amount of data. The visual information present in the training data can be misused, which creates a huge privacy concern. One of the prominent solutions for this issue…

Artificial neural networks (ANNs) have gained significant popularity in the last decade for solving narrow AI problems in domains such as healthcare, transportation, and defense. As ANNs become more ubiquitous, it is imperative to…

Together with impressive advances touching every aspect of our society, AI technology based on Deep Neural Networks (DNN) is bringing increasing security concerns. While attacks operating at test time have monopolised the initial attention…

In the exciting generative AI era, the diffusion model has emerged as a very powerful and widely adopted content generation and editing tool for various data modalities, making the study of their potential security risks very necessary and…

Recent studies have shown that sponge attacks can significantly increase the energy consumption and inference latency of deep neural networks (DNNs). However, prior work has focused primarily on computer vision and natural language…